A look back at an intense year in intensive care

Three experts analyze the pandemic response.


When the Society of Critical Care Medicine brought three experts together in February to discuss COVID-19 care, especially oxygenation, a major focus was the pace of change over the past year.

“We were very, very conservative at the beginning. I felt like I was dressing to go to war every time I went in, and so that's changed a lot,” said Hayley B. Gershengorn, MD, an intensivist and associate professor at the University of Miami. “Staff at the hospital had a very hard time with [rapid changes]. And the biggest example of that would be when we changed what PPE we thought was appropriate.”

Image by Getty Images
Image by Getty Images

Sudden shifts in recommendations, including about personal protective equipment, were disconcerting “even for me to think about it,” agreed Roshni S. Sreedharan, MD, an anesthesiologist intensivist at the Cleveland Clinic in Ohio. “Published in a very reputable journal, we have a study that may not hold in the next week or the next month. But we are going with what we got.”

The physicians discussed the evolution of practice and protocols for COVID-19, including ongoing challenges and possible missteps, along with Bram Rochwerg, MD, MSc, an associate professor at McMaster University in Hamilton, Canada, during a “cross-talk” session of the 2021 Critical Care Congress.

The preferred means for providing oxygen to patients with COVID-19 has been one of the most significant areas of change, all agreed.

“So much fear factored in early on in this pandemic, especially around noninvasive strategies...with the uncertainty around how this affects aerosolization and risk to health care workers,” said Dr. Rochwerg. “We were jumping right to intubation in some of these patients that were tenuous, and I think it was well intentioned, but now looking back in retrospect, perhaps the wrong thing to do.”

Dr. Gershengorn agreed, and she took responsibility for this choice at her hospital. “I own our protocol. I was the one who drafted it. It said that if they require six liters or more of low-flow nasal cannula oxygen or require more than 50% [by] face mask, we should be considering early intubation.”

The Cleveland Clinic had a similar policy, according to Dr. Sreedharan. “That kind of also inculcated fear in the nurses and the [respiratory therapists] because they knew that it was on the protocol. And if a physician was saying, ‘No, let's just watch this patient clinically, they kind of look OK,’ there was fear, there was apprehension,” she said. “But what I'm curious about is, why do you think that came about—the six liters?”

Dr. Rochwerg responded. “It seems relatively arbitrary, to be honest,” he said. Intubation may have been used in cases when high-flow oxygen, in particular, would have been appropriate, he suggested. “The introduction of high flow in the last five or 10 years in adult critical care, I'm convinced, has saved so many people from being intubated, and COVID isn't, I don't think, any different.”

COVID-19 was different, of course, in its infection control aspects, and that drove initial protocols regarding high-flow nasal cannula and noninvasive ventilation (NIV). “It was an absolute no, in the beginning,” said Dr. Sreedharan. “It was the apprehension as to, ‘Is this airborne or not?’”

Dr. Rochwerg reported that he and some colleagues searched the literature for research on aerosolization with high-flow oxygen. “There's no data. Everything seems much more expert-driven and very subjective,” he said.

The initial caution was understandable, he added. “We all want to make sure that we're protecting health care workers. But at the same time, I think you can go too far at the risk of detriment to your patients, and we know the harms of pulling the trigger on intubation too early,” said Dr. Rochwerg. “Those first few weeks, we were ignoring noninvasive tools, which are so crucial for keeping patients off the ventilator.”

Practice has definitely changed since, according to Dr. Gershengorn. “Now as a reaction to that, I feel like we've even gone a little bit too far in the opposite direction,” she said. “We've become incredibly comfortable, or I've become incredibly comfortable, watching people really hover in [an oxygen saturation] range that I'm not so sure is so great.”

Dr. Sreedharan seconded that observation and pointed out one potential cause of physicians' reluctance to intubate. “I noticed that the COVID patients who were on high flow, were talking, and looked comfortable. They were profoundly hypoxic, but they looked different from our other regular pneumonia patients,” she said.

That could be because COVID-19 patients have been younger and healthier than the typical ICU patient with pneumonia, speculated Dr. Gershengorn. “For me, that's been really challenging, because it's very hard to convince that person that we think it's in their best interest to be sedated and put on a ventilator,” she said. “Especially because so many people at this point have heard so much about what they interpret to be their outcomes if they are placed on a ventilator.”

But there are hazards with the alternatives, too. “I get so nervous because patients can look so good on high flow,” Dr. Rochwerg said. “If that mask falls off, or they pull it off, they can desat and have a cardiac arrest. It wouldn't take long at all.”

Clinicians should carefully consider the risks when choosing high-flow oxygen or NIV over intubation, the experts said. “I fear that what was meant as an ‘It's OK from a safety perspective’ became an ‘It's OK, and maybe even correct, from a clinical care perspective,’” said Dr. Gershengorn. “We are using, in my opinion, way too much NIV.”

“I 100% agree,” said Dr. Sreedharan. “If we had to put somebody on NIV for 24 hours, we probably should be intubating this person...because their mouth is dry, and they're not able to talk, and they cannot eat, and it's just a snowballing thing that goes on, which we would be very wary about if it was not COVID.”

Decision making about NIV can be particularly challenging in patients with diagnosed or suspected COVID-19 and comorbidities such as chronic obstructive pulmonary disease or heart failure. “You want to use NIV because these are conditions that you know respond quickly,” said Dr. Rochwerg. “But when you're concerned enough about COVID, those that come in with hypoxemic respiratory failure, pneumonia, these patients don't do well with NIV.”

One traditional downside of NIV is its effects on patients' faces, including skin ulcers, so Dr. Rochwerg was very excited to talk about a solution newly available at his hospital. “We've just received 100 NIV helmets in the last two weeks, and we've just used our first helmet interface on a COVID patient.”

NIV provided through a helmet has shown success in other countries, including Italy, during the pandemic, Dr. Rochwerg reported. “Not just in the ICU, but they were using it on the ward because the risks are way less. Even if the patient vomits, it just collects at the bottom of the helmet. The risk for aspiration is much lower,” he said.

Drs. Sreedharan and Gershengorn agreed that helmets seem like promisingly technology but said that neither of their hospitals have them.

Even without novel technology, there have been some shifting standards about what severity of illness necessitates ICU admission, the experts said. “After COVID, things have changed. The tolerance has changed as to how much [need for oxygen] is OK on the floor versus what is not,” said Dr. Sreedharan.

“At our institution, we've allowed up until 50% FiO2 high flow on the ward,” said Dr. Rochwerg. “I do worry when our comfort of keeping patients on the ward goes up that some of these patients get left there too long, especially those on high flow.”

At the University of Miami, the cohorting of all COVID-19 patients on a single floor helped alleviate that concern, Dr. Gershengorn said. “There's a comfort in having those people on a non-ICU team, because they're kind of in the same place. There are ICU nurses around and the critical care team is always on that floor, even if they're not under our care.”

There have, however, been changes in the availability of subspecialty expertise for these patients, she noted. “Traditionally, all of our non-ICU patients who require either high flow or some sort of noninvasive therapy had a pulmonary consultation,” said Dr. Gershengorn. “Partially because our group has been pretty stretched, like has been true in many places, we've pulled back a bit on that.”

Surges have been a key determinant of who gets intensive care, agreed Dr. Rochwerg. “So much of this depends on the current pressures that the hospital is facing, and if we say that it doesn't, we're lying. Even on a week-by-week, day-by-day basis, if I have beds, my threshold for admitting to the ICU is lower.”

It's not just about beds, noted Dr. Gershengorn. “An ICU bed is just a regular bedroom bed, with an ICU team—I think that that's the more important thing,” she said. She noted that her institution was fortunate to be able to hire temporary contract employees to deal with the surge in critically ill patients.

“So we actually have a lot of people, specifically nurses and respiratory therapists, who work around the country, right, and so they bring with them perspective from many different places during COVID,” Dr. Gershengorn said. “That's been really interesting, actually, because I think that's allowed for a lot more comfort with things being a little different.”

For most clinicians, the pace of change has been more challenging. “It is tough, especially for people that are used to protocols that stay the same, you know, for five years, 10 years at a time,” said Dr. Rochwerg. “Over the last 10 months, something is only good for the day that it was made, and quite often things change day by day, hour by hour.”

That dynamic adds a cognitive load in these already stressful times that really can't be fully ameliorated. “Honest communication, I think, is the best thing you can do,” said Dr. Rochwerg. “Reminding them that, you know, we're doing this based on the best data that we have. And we might have different data tomorrow and might have to make different decisions.”

Dr. Sreedharan agreed that transparency is the best solution until there can be truly definitive evidence. “Just saying that ‘We are all in this together. We are trying to get all the information that we can and trying to work with the best stuff we got,’” she said. “Ten years from now, we'll look back at this and think about everything that we did, and dissect it in detail, but right now, it's such an evolving process.”